CN109048924A - A kind of intelligent robot flexible job devices and methods therefor based on machine learning - Google Patents
A kind of intelligent robot flexible job devices and methods therefor based on machine learning Download PDFInfo
- Publication number
- CN109048924A CN109048924A CN201811232551.5A CN201811232551A CN109048924A CN 109048924 A CN109048924 A CN 109048924A CN 201811232551 A CN201811232551 A CN 201811232551A CN 109048924 A CN109048924 A CN 109048924A
- Authority
- CN
- China
- Prior art keywords
- module
- robot
- job
- data
- training
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1602—Programme controls characterised by the control system, structure, architecture
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1628—Programme controls characterised by the control loop
- B25J9/163—Programme controls characterised by the control loop learning, adaptive, model based, rule based expert control
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1694—Programme controls characterised by use of sensors other than normal servo-feedback from position, speed or acceleration sensors, perception control, multi-sensor controlled systems, sensor fusion
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1694—Programme controls characterised by use of sensors other than normal servo-feedback from position, speed or acceleration sensors, perception control, multi-sensor controlled systems, sensor fusion
- B25J9/1697—Vision controlled systems
Abstract
The invention discloses a kind of intelligent robot flexible job devices and methods therefor based on machine learning, data acquisition module multi-angle multi-pose carries out data acquisition to the standard manufacturing process of product, data scaling and training module connect training and study for production operation flow process with data acquisition module, job analysis module is connect for providing motion planning model for robot control with data scaling and training module, control execution module is connect with job analysis module carries out movement execution for controlling intelligent robot, mechanics sensing module is connect for acquiring to the power real-time perfoming in implementation procedure with job analysis module.The intelligent flexible that the present invention realizes robot is turned into industry.
Description
Technical field
The present invention relates to a kind of robot manipulating task devices and methods therefor, especially a kind of intelligence machine based on machine learning
People's flexible job devices and methods therefor.
Background technique
With industry 4.0 and the proposition of China 2025, factory automation and production it is intelligent with intelligent production line,
Robot, software have close association, how to realize that intelligent, wisdom generation is currently manufacturing enterprise, robot
Manufacturer, research institution and colleges and universities are all the problem of studying, and the application of robot is focused primarily upon according to specific program at present
Execute the movement of repeatability, or the movements such as simple sorting, loading and unloading, assembly realized by the positioning of vision, with intelligence,
The production of wisdom has very big gap.And programming for robot and the familiar of production technology have very big difference,
It is unable to meet production the flexible job requirement of type enterprise.Therefore a kind of robot flexibility operational method and dress based on machine learning
Setting has great significance to multitask, complicated flexible job.
Summary of the invention
Technical problem to be solved by the invention is to provide a kind of, and the intelligent robot flexible job based on machine learning fills
It sets and its method, realizes that the intelligent flexible of robot is turned into industry.
In order to solve the above technical problems, the technical scheme adopted by the invention is that:
A kind of intelligent robot flexible job device based on machine learning, it is characterised in that: comprising data acquisition module,
Data scaling and training module, job analysis module, control execution module and mechanics sensing module；Data acquisition module multi-angle
Multi-pose carries out data acquisition to the standard manufacturing process of product, and data scaling and training module connect use with data acquisition module
In the training and study of production operation flow process, job analysis module is connect for being robot control with data scaling and training module
System provide motion planning model, control execution module connect with job analysis module for control intelligent robot carry out act hold
Row, mechanics sensing module are connect for acquiring to the power real-time perfoming in implementation procedure with job analysis module.
Further, the data acquisition module is arranged in intelligent robot front end, is in real time transmitted data by agreement
The training and study of production operation flow process are carried out to data scaling and training module.
Further, the data acquisition module include with video acquisition function sensor, and to temporal information into
Row record, forms complete video flowing and is transmitted, while carrying out during machine learning multi-faceted multiple to standard
Operation carries out record acquisition, carries out transmission control to data by communications protocol.
Further, the mechanics sensing module is arranged in intelligent robot executing agency front end, in implementation procedure
The acquisition of power real-time perfoming, and the real-time data transmission of acquisition to job analysis module is subjected to data analytical judgment, it is robot
Correct operation provides feedback information, when power is more than the threshold value of setting, optimizes robot manipulating task posture or stops robot fortune
Row.
A kind of flexible job method of the intelligent robot flexible job device based on machine learning, it is characterised in that include
Following steps:
Step 1: data acquisition module multi-angle multi-pose carries out data acquisition to the standard manufacturing process of product；
Step 2: data scaling and training module carry out analysis and instruction in image processor to the extraction of data characteristics
Practice, generates the type and record production process of feature object；
Step 3: mechanics sensing module is acquired transmission to the mechanics in operation process and mentions to the operation of intelligent robot
The closed loop feedback of force prevents the excessive damage product of the pressure of robot manipulating task；
Step 4: job analysis module reads the information flow of data scaling and training module, in conjunction with mechanics sensing module
Feedback, the machine instruction for forming closed-loop control are transmitted in Study of Intelligent Robot Control system and carry out work flow control；
Step 5: control execution module generates instruction and operates to production operation.
Further, the step 1 is specially that data acquisition module passes through video acquisition sensor to skilled production work
The standard operation for making personnel carries out multi-faceted video data acquiring, record, and forms video flowing and be uploaded to image analysis system,
The training dataset of machine learning is carried out for data scaling and training module.
Further, the step 4 is specially
4.1 generate job instruction model；
4.2 generate work flow list；
4.3 poses determine adjustment actuating mechanism posture；
4.4 executing agency's speed plannings and judgement；
4.5 executing agency's mechanics feedback and judgement；
4.6 current tasks execute and return to 4.2.
Further, the generation work flow list be specially to data scaling and training module generate job information into
Row sorts out optimization, and carries out classification processing according to K-means and SVM algorithm, carries out clustering processing, and according to time and efficiency
Optimal theory provides the job list of optimization.
Further, the pose determines adjustment actuating mechanism posture and executing agency's speed planning and determines to be specially machine
Device people carries out crawl operation to raw material according to work flow, optimizes the speed control method of sine wave, drop in the process of grasping
Rigidity in low operation process carries out low speed when close to target object and approaches, and accurate adjustment actuating mechanism carries out the essence of operation
Degree and flexibility, avoid to surrounding and execute the damage of article.
Further, executing agency's mechanics feedback and judgement are specially during robot manipulating task, using three
The information processing of closed loop, using method for optimally controlling, according to the analysis of power in robot implementation procedure, with the power in statistical value
It is compared, determines the state of current robot actuating mechanism, if power is more than threshold value 1, reverse adjustment executing agency state, if power
Greater than threshold value 2, then stop robot.And job parameter is adjusted, carry out the planning again in path.
Compared with prior art, the present invention having the following advantages that and effect:
1, the present invention carries out the calibration and extraction of feature to the information of acquisition, and the mean value of optimization is used to the frame information of video
Filtering algorithm is filtered, and carries out classification calibration, shape to the build-up member of product and using tool types using SVM algorithm
At components and tool table listings it is more clear accurate；
2, the present invention uses LSTM neural network, and the interval delay in processing and predicted time sequence that is more suitable is relatively long
Information, to be formed in temporal flow chart；
3, the present invention uses power, speed, the Study of Intelligent Robot Control system of position tricyclic closed feedback, all according to system
Work flow is chosen according to the confidence level to component, during operation due to component put be not it is carefully and neatly done,
System carries out accurate adjustment to robot front end using Adaptive Fuzzy Control, reaches accurately controlling for efficiency and precision.
Detailed description of the invention
Fig. 1 is a kind of schematic diagram of intelligent robot flexible job device based on machine learning of the invention.
Fig. 2 is a kind of flow chart of intelligent robot flexible job method based on machine learning of the invention.
Specific embodiment
Below by specific embodiment, the present invention is described further, and following embodiment is explanation of the invention
And the invention is not limited to following embodiments.
As shown in Figure 1, a kind of intelligent robot flexible job device based on machine learning of the invention is adopted comprising data
Collect module, data scaling and training module, job analysis module, control execution module and mechanics sensing module；Data acquisition module
Block multi-angle multi-pose carries out data acquisition, data scaling and training module and data acquisition module to the standard manufacturing process of product
Block connection be used for production operation flow process training and study, job analysis module connect with data scaling and training module for for
Robot control provide motion planning model, control execution module connect with job analysis module for control intelligent robot into
Action executes, and mechanics sensing module is connect for acquiring to the power real-time perfoming in implementation procedure with job analysis module.
Data acquisition module setting sends data to data scaling and instruction by agreement in real time in intelligent robot front end
Practice training and study that module carries out production operation flow process.Data acquisition module include video acquisition function or have it is similar
The sensor of function, and temporal information can be recorded, it forms complete video flowing and is transmitted, while in machine learning
Carry out in the process it is multi-faceted it is multiple record acquisition is carried out to standard operation, transmission control is carried out to data by communications protocol.
To the transmission information of data acquisition module, the calibration and extraction of feature are carried out, optimization is used to the frame information of video
Mean Filtering Algorithm be filtered, classified using SVM algorithm to the build-up member of product and using tool types
Calibration.Form the table listings of components and tool.Simultaneously because needing to form temporal flow chart, therefore use LSTM
Neural network, it is advantageous that the information that interval delay is relatively long in processing and predicted time sequence is more suitable, thus shape
At flow chart in time.
Data scaling and training module are used to carry out the extraction of data characteristics the analysis and training in image processor, raw
At the type and record production process of feature object.
Job analysis module is used to read the information flow of data scaling and training module, in conjunction with the anti-of mechanics sensing module
Feedback, the machine instruction for forming closed-loop control are transmitted in Study of Intelligent Robot Control system and carry out work flow control.
Mechanics sensing module is arranged in intelligent robot executing agency front end, adopts to the power real-time perfoming in implementation procedure
Collection, and the real-time data transmission of acquisition to job analysis module is subjected to data analytical judgment, it is provided for the correct operation of robot
Feedback information optimizes robot manipulating task posture or stops robot operation when power is more than the threshold value of setting.Mechanics perceives mould
Block is used to be acquired the mechanics in operation process transmission and provides the closed loop feedback of power to the operation of intelligent robot, prevents machine
The excessive damage product of the pressure of device people's operation.
A kind of flexible job method of the intelligent robot flexible job device based on machine learning comprising the steps of:
Step 1: data acquisition module multi-angle multi-pose carries out data acquisition to the standard manufacturing process of product；
Data acquisition module carries out the standard operation of skilled production work personnel by video acquisition sensor multi-party
Video data acquiring, the record of position, and form video flowing and be uploaded to image analysis system, it is carried out for data scaling and training module
The training dataset of machine learning.
Step 2: data scaling and training module carry out analysis and instruction in image processor to the extraction of data characteristics
Practice, generates the type and record production process of feature object；
Step 3: mechanics sensing module is acquired transmission to the mechanics in operation process and mentions to the operation of intelligent robot
The closed loop feedback of force prevents the excessive damage product of the pressure of robot manipulating task；
It is all according to system job stream using power, speed, the Study of Intelligent Robot Control system of position tricyclic closed feedback
Journey is chosen according to the confidence level to component, since putting for component is not carefully and neatly that system is adopted during operation
Accurate adjustment is carried out to robot front end with Adaptive Fuzzy Control, reaches accurately controlling for efficiency and precision.
Step 4: job analysis module reads the information flow of data scaling and training module, in conjunction with mechanics sensing module
Feedback, the machine instruction for forming closed-loop control are transmitted in Study of Intelligent Robot Control system and carry out work flow control；
Specially
4.1 generate job instruction model；
4.2 generate work flow list；Job information is generated to data scaling and training module and carries out classification optimization, and root
Classification processing is carried out according to K-means and SVM algorithm, carries out clustering processing, and according to the theory of time and efficiency optimization, is provided
The job list of optimization.
4.3 poses determine adjustment actuating mechanism posture；
4.4 executing agency's speed plannings and judgement；Robot carries out crawl operation to raw material according to work flow, is grabbing
The speed control method for optimizing sine wave during taking, reduces the rigidity in operation process, carries out when close to target object low
Quick access is close, and accurate adjustment actuating mechanism carries out the precision and flexibility of operation, avoids to surrounding and execute the damage of article.
4.5 executing agency's mechanics feedback and judgement；During robot manipulating task, using the information processing of three closed loops,
It using method for optimally controlling, according to the analysis of power in robot implementation procedure, is compared with the power in statistical value, determines mesh
The state of preceding robot actuating mechanism, if power is more than threshold value 1, reverse adjustment executing agency state is stopped if power is greater than threshold value 2
Only robot.And job parameter is adjusted, carry out the planning again in path.
4.6 current tasks execute and return to 4.2.It has executed and parameter is subjected to record storage in the number of system after current task
According to concentration, increase the self study data set of intelligent robot, and provides foundation for the adjustment of lower subparameter, while returning to execution task
4.2, until task is completed.
Step 5: control execution module generates instruction and operates to production operation.
Above content is only illustrations made for the present invention described in this specification.Technology belonging to the present invention
The technical staff in field can make various modifications or additions to the described embodiments or by a similar method
Substitution, content without departing from description of the invention or beyond the scope defined by this claim should belong to this
The protection scope of invention.
Claims (10)
1. a kind of intelligent robot flexible job device based on machine learning, it is characterised in that: include data acquisition module, number
According to calibration and training module, job analysis module, control execution module and mechanics sensing module；Data acquisition module multi-angle is more
Posture carries out data acquisition to the standard manufacturing process of product, and data scaling and training module connect with data acquisition module and be used for
The training and study of production operation flow process, job analysis module are connect for controlling for robot with data scaling and training module
There is provided motion planning model, control execution module connect with job analysis module for control intelligent robot carry out act hold
Row, mechanics sensing module are connect for acquiring to the power real-time perfoming in implementation procedure with job analysis module.
2. a kind of intelligent robot flexible job device based on machine learning described in accordance with the claim 1, it is characterised in that:
The data acquisition module setting sends data to data scaling and training mould by agreement in real time in intelligent robot front end
The training and study of block progress production operation flow process.
3. a kind of intelligent robot flexible job device based on machine learning described in accordance with the claim 1, it is characterised in that:
The data acquisition module includes the sensor with video acquisition function, and is recorded to temporal information, is formed complete
Video flowing is transmitted, at the same carry out during machine learning it is multi-faceted it is multiple record acquisition is carried out to standard operation,
Transmission control is carried out to data by communications protocol.
4. a kind of intelligent robot flexible job device based on machine learning described in accordance with the claim 1, it is characterised in that:
The mechanics sensing module setting acquires the power real-time perfoming in implementation procedure in intelligent robot executing agency front end, and
The real-time data transmission of acquisition to job analysis module is subjected to data analytical judgment, provides feedback letter for the correct operation of robot
Breath optimizes robot manipulating task posture or stops robot operation when power is more than the threshold value of setting.
5. a kind of the flexible of the described in any item intelligent robot flexible job devices based on machine learning of claim 1-4 is made
Industry method, it is characterised in that comprise the steps of:
Step 1: data acquisition module multi-angle multi-pose carries out data acquisition to the standard manufacturing process of product；
Step 2: data scaling and training module carry out analysis and training in image processor to the extraction of data characteristics, raw
At the type and record production process of feature object；
Step 3: mechanics sensing module is acquired transmission to the mechanics in operation process and provides power to the operation of intelligent robot
Closed loop feedback, prevent the excessive damage product of the pressure of robot manipulating task；
Step 4: job analysis module reads the information flow of data scaling and training module, in conjunction with the feedback of mechanics sensing module,
The machine instruction for forming closed-loop control is transmitted in Study of Intelligent Robot Control system and carries out work flow control；
Step 5: control execution module generates instruction and operates to production operation.
6. flexible job method according to claim 5, it is characterised in that: the step 1 is specially data acquisition module
Multi-faceted video data acquiring, note are carried out to the standard operation of skilled production work personnel by video acquisition sensor
Record, and form video flowing and be uploaded to image analysis system, the training data of machine learning is carried out for data scaling and training module
Collection.
7. flexible job method according to claim 5, it is characterised in that: the step 4 is specially
4.1 generate job instruction model；
4.2 generate work flow list；
4.3 poses determine adjustment actuating mechanism posture；
4.4 executing agency's speed plannings and judgement；
4.5 executing agency's mechanics feedback and judgement；
4.6 current tasks execute and return to 4.2.
8. flexible job method according to claim 7, it is characterised in that: the generation work flow list is specially pair
Data scaling and training module generate job information and carry out classification optimization, and are carried out at classification according to K-means and SVM algorithm
Reason carries out clustering processing, and according to the theory of time and efficiency optimization, provides the job list of optimization.
9. flexible job method according to claim 7, it is characterised in that: the pose determines adjustment actuating mechanism posture
Crawl operation is carried out to raw material according to work flow with executing agency's speed planning and judgement specially robot, was being grabbed
The speed control method for optimizing sine wave in journey, reduces the rigidity in operation process, and low speed is carried out when close to target object and is connect
Closely, accurate adjustment actuating mechanism carries out the precision and flexibility of operation, avoids to surrounding and execute the damage of article.
10. flexible job method according to claim 7, it is characterised in that: executing agency's mechanics feedback and judgement
Specially during robot manipulating task, executed using method for optimally controlling in robot using the information processing of three closed loops
It in the process according to the analysis of power, is compared with the power in statistical value, determines the state of current robot actuating mechanism, if power is super
Threshold value 1 is crossed, reverse adjustment executing agency state stops robot if power is greater than threshold value 2.And job parameter is adjusted, carry out road
The planning again of diameter.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811232551.5A CN109048924A (en) | 2018-10-22 | 2018-10-22 | A kind of intelligent robot flexible job devices and methods therefor based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811232551.5A CN109048924A (en) | 2018-10-22 | 2018-10-22 | A kind of intelligent robot flexible job devices and methods therefor based on machine learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109048924A true CN109048924A (en) | 2018-12-21 |
Family
ID=64765253
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811232551.5A Pending CN109048924A (en) | 2018-10-22 | 2018-10-22 | A kind of intelligent robot flexible job devices and methods therefor based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109048924A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109591013A (en) * | 2018-12-12 | 2019-04-09 | 山东大学 | A kind of flexible assembly analogue system and its implementation |
| CN112101145A (en) * | 2020-08-28 | 2020-12-18 | 西北工业大学 | SVM classifier based pose estimation method for mobile robot |
| US11745338B2 (en) | 2020-03-13 | 2023-09-05 | Omron Corporation | Control apparatus, robot, learning apparatus, robot system, and method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106335057A (en) * | 2016-09-27 | 2017-01-18 | 东南大学 | Total-space smooth hole insertion control method applied to assembly robot and based on real-time force control |
| CN107127735A (en) * | 2016-02-29 | 2017-09-05 | 通用汽车环球科技运作有限责任公司 | People's demonstration formula has the robot learning of power and position purpose task |
| US20180056434A1 (en) * | 2016-08-30 | 2018-03-01 | Fanuc Corporation | Spot welding apparatus that judges welding state |
| CN108241296A (en) * | 2016-12-26 | 2018-07-03 | 发那科株式会社 | Learn the machine learning device and component assembly system of Assembly Action |
| EP3357649A2 (en) * | 2017-02-06 | 2018-08-08 | Seiko Epson Corporation | Control device, robot, and robot system |
| CN108519814A (en) * | 2018-03-21 | 2018-09-11 | 北京科技大学 | A kind of man-machine interactive operation system |
-
2018
- 2018-10-22 CN CN201811232551.5A patent/CN109048924A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107127735A (en) * | 2016-02-29 | 2017-09-05 | 通用汽车环球科技运作有限责任公司 | People's demonstration formula has the robot learning of power and position purpose task |
| US20180056434A1 (en) * | 2016-08-30 | 2018-03-01 | Fanuc Corporation | Spot welding apparatus that judges welding state |
| CN106335057A (en) * | 2016-09-27 | 2017-01-18 | 东南大学 | Total-space smooth hole insertion control method applied to assembly robot and based on real-time force control |
| CN108241296A (en) * | 2016-12-26 | 2018-07-03 | 发那科株式会社 | Learn the machine learning device and component assembly system of Assembly Action |
| EP3357649A2 (en) * | 2017-02-06 | 2018-08-08 | Seiko Epson Corporation | Control device, robot, and robot system |
| CN108519814A (en) * | 2018-03-21 | 2018-09-11 | 北京科技大学 | A kind of man-machine interactive operation system |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109591013A (en) * | 2018-12-12 | 2019-04-09 | 山东大学 | A kind of flexible assembly analogue system and its implementation |
| US11745338B2 (en) | 2020-03-13 | 2023-09-05 | Omron Corporation | Control apparatus, robot, learning apparatus, robot system, and method |
| TWI827907B (en) * | 2020-03-13 | 2024-01-01 | 日商歐姆龍股份有限公司 | Robot control devices, control methods, robots and their systems, learning devices, methods and computer program products |
| CN112101145A (en) * | 2020-08-28 | 2020-12-18 | 西北工业大学 | SVM classifier based pose estimation method for mobile robot |
| CN112101145B (en) * | 2020-08-28 | 2022-05-17 | 西北工业大学 | SVM classifier based pose estimation method for mobile robot |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106737673B (en) | A method of the control of mechanical arm end to end based on deep learning | |
| CN107598928B (en) | Camera and robot control system and its automatic adaptation method based on semantic model | |
| CN109048924A (en) | A kind of intelligent robot flexible job devices and methods therefor based on machine learning | |
| Niekum et al. | Learning and generalization of complex tasks from unstructured demonstrations | |
| CN110125930A (en) | It is a kind of that control method is grabbed based on the mechanical arm of machine vision and deep learning | |
| DE102012213188B4 (en) | A method and system for controlling an execution sequence of a skilled robot using a condition classification | |
| CN111251277B (en) | Human-computer collaboration tool submission system and method based on teaching learning | |
| Wachs et al. | Real-time hand gesture telerobotic system using fuzzy c-means clustering | |
| CN108672316A (en) | A kind of micro parts quality detecting system based on convolutional neural networks | |
| Fu et al. | Active learning-based grasp for accurate industrial manipulation | |
| Miyajima | Deep learning triggers a new era in industrial robotics | |
| CN108818537A (en) | A kind of robot industry method for sorting based on cloud deep learning | |
| CN110363214A (en) | A kind of contact condition recognition methods of the robotic asssembly based on GWA-SVM | |
| Yu-ming et al. | Research on intelligent manufacturing flexible production line system based on digital twin | |
| Liu et al. | A mixed perception-based human-robot collaborative maintenance approach driven by augmented reality and online deep reinforcement learning | |
| CN116442219B (en) | Intelligent robot control system and method | |
| Yang | Research on grasping method of industrial robot based on deep learning and machine vision | |
| CN113311789A (en) | Control method of warehousing robot based on 5G and brain-like pulse neural network model | |
| Yang et al. | Domain centralization and cross-modal reinforcement learning for vision-based robotic manipulation | |
| Sawaragi et al. | Self-reflective segmentation of human bodily motions using recurrent neural networks | |
| Ovchar et al. | Automated recognition and sorting of agricultural objects using multi-agent approach | |
| CN206200972U (en) | It is a kind of by catching hand exercise come the system of auxiliary mechanical arm | |
| CN112257655B (en) | Method for robot to recognize human body sewing action | |
| Man et al. | Robot bolt skill learning based on GMM-GMR | |
| Wang | Application and Development of Neural Network Technology in Mechanical Automation Processing Parameters |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181221 |
|
| RJ01 | Rejection of invention patent application after publication |